How Do You Choose a Tempo That Fits the Work and Remains Controllable?

Building a Piano Piece · Season Three: From Score to Interpretation: Building a Performance of Your Own · Article Two

Tempo choice is often reduced to a number: how many beats per minute should this piece go? If a metronome mark is printed, the answer appears to have been supplied. If the score offers only Allegro, Andante or Adagio, a player may sample several recordings and adopt a preferred speed.

The real task is more demanding. A tempo word, metre, beat unit, shortest note values, harmonic rhythm, phrase length, texture, instrument and room jointly shape what speed means. The performer’s movement is not an irrelevant external limitation. A tempo that survives the opening but destroys voicing, articulation or quiet control in the dense passage has not yet become a viable tempo for the work.

The useful aim is therefore not an isolated correct number. It is a tempo corridor: a range in which the work’s character, structural intelligibility and the performer’s control remain simultaneously available. Within that range, the pianist can choose a reference pulse and design local flexibility.

What exactly does the number measure?

A metronome mark has meaning only with its beat unit. A crotchet at 120 and a minim at 120 are not neighbouring interpretations. In compound metre, the player must also decide whether the felt motion lies in subdivisions or larger beats. A historical metronome figure raises a source question: was it authorised by the composer, added by an editor, inferred later, or transmitted in one of several conflicting sources?

The continuing discussion of Beethoven’s metronome marks demonstrates why a number cannot be detached from its history. Early instruments, the composer’s marks, revisions and later performance traditions do not merge into one uncomplicated instruction. The Beethoven-Haus account of the metronome supplies useful historical context. Research comparing Beethoven tempo indications and recordings also shows that the relation must be examined work by work, not replaced by slogans claiming that all modern performances are too slow or that every printed figure must be obeyed identically.

Tempo words are not merely numerical bands either. Allegro can imply activity or forward character; grave concerns weight and seriousness as well as pace. Metre and grouping alter the felt motion at the same BPM. Counting four small beats and sensing two larger beats may consume the same clock time while producing different phrase gravity.

Find the work’s structural lower boundary

The lower boundary is not the slowest speed at which the notes can still be played. It is the point below which relationships within the work begin to disintegrate.

Ask four questions. Does a long melody still register as one utterance? Does harmonic suspense remain active across neighbouring events? Does a dance or travelling rhythm retain motion? Do ornaments and short values belong to a main line, or have they become isolated objects? If a phrase turns into a sequence of waits, or every bar sounds like a fresh beginning, the pace may be spacious but the syntax has weakened.

Do not test only the most difficult bar. Choose a long phrase, a transition and a closing passage. Record three progressively slower versions of each. Listen later without looking at the numbers and note where direction first disappears. At what point do you compensate with exaggerated dynamics because motion has gone? The vicinity of that failure is the structural lower boundary.

Find the performer’s controllable upper boundary

The upper boundary is not the fastest accident-free attempt. At a controllable tempo, pitch and rhythm remain dependable; important voices can still be differentiated; articulation does not become involuntarily uniform; soft playing reaches the key bed; pedalling follows harmony; the performer can recover from a small disruption; and the ending does not accelerate or collapse through fatigue.

Experimental studies support a modest but important conclusion. Increasing rate changes timing variability, key velocity and finger movement; it does not merely compress an otherwise identical action. Research on tempo, keystroke velocity and finger movement and on temporal control by skilled pianists at different rates cannot prescribe a sonata’s tempo, but it does show that tempo forms part of motor organisation.

Test the densest place, quietest place, largest displacement and ending. Play each checkpoint three times, then restore at least a phrase before and after it. Success in an isolated cell is not yet evidence that the whole piece can carry that baseline speed.

Draw three boundaries, not one target

The practice record can contain three lines.

The structural lower boundary is the point below which line, dance or harmonic direction becomes obscure. The motor upper boundary is the point above which tone, voicing, articulation or recovery deteriorates. Between them lies an expressively viable zone: the part of the range in which character and proportion emerge without constant rhetorical repair.

Choose three reference tempi separated by roughly five to eight per cent and label them A, B and C. Record whole sections, not openings alone. A day later, randomise playback and score pulse, phrase, harmony, articulation, voicing, bodily strain and ending stability. Do not ask only which version you like. State what each version loses.

Suppose A is structurally lucid but loses impetus, while C is exciting but erases inner voices and quiet articulation. B may be the current choice. “Current” matters: the motor upper boundary can move as control improves, and a more brilliant piano or resonant room may shift the viable zone. The structural lower boundary, however, does not yield merely because a performer wants more display.

Local timing is not the baseline tempo

Once a reference tempo is established, the performer still decides how cadences, modulations, climaxes and new sections depart from it. Local slowing should not conceal a pulse that was never stable. First play a relatively even baseline and confirm that structure remains audible without rubato. Then introduce departures for explicit reasons.

Use a return test. Allow time before a cadence; after the new phrase begins, do body and ear know which pulse to resume? If each attempt returns at a different rate, the apparent freedom may simply reveal an undefined baseline. A second experiment keeps an accompaniment or inner voice relatively referential while allowing small melodic flexibility, then permits the entire texture to move together. Compare the structural and stylistic results. Neither solution is a universal rule.

Make a tempo-corridor record

For the piece being studied, record:

1. the tempo word, metre and provenance of any metronome mark;
2. the unit in which pulse is felt, with a reason;
3. the structural lower boundary and the first relationship that fails;
4. the motor upper boundary and the first control that fails;
5. BPM and listening notes for three complete recordings;
6. the selected baseline and permitted local departures;
7. results of cold starts, endings and tests on another piano.

Finish with a cold run on another day. Do not pre-play the difficult spots. Record from beginning to end. If the chosen tempo exists only after extensive warm-up, immediately after drilling the hardest passage, or under unsustainable arousal, it is not yet a tempo the performer can publicly carry.

A suitable tempo is neither the average of famous recordings nor the highest speed briefly accessible to the fingers. It is a range in which the work’s relationships remain audible, interpretive choices remain available and the performer can take responsibility from first note to last.

Tempo is not a number attached to a work. It is a testable agreement between the work’s structure and the performer’s present control.

Primary sources and further reading

• Beethoven-Haus: Beethoven and the metronome
• Research on Beethoven tempo indications, recordings and tempo models
• Effects of tempo on piano timing, key velocity and finger movement
• Temporal control by skilled pianists at different tempi
• A framework for timing, dynamics and articulation in expressive performance

Continue reading: Building a Piano Piece series page

If you would like to bring these ideas about listening, understanding, and practice to the keyboard, you might try ScoreFlow, an app I developed to make score reading and daily practice flow more naturally together.